Tele-Operated Driving
Current automated driving vehicle prototypes prove the feasibility of truly driverless cars. Tele-operated Driving (ToD) can be leveraged as an enabling technology to smooth this transition, as edge cases remain, which necessitates falling back on human operators. For ToD, an interface over the mobile 5G network is created that allows a human to remotely control a vehicle. Through such an interface, sensor and vehicle data, e.g., video feeds and velocity are transmitted from the vehicle to the vehicle control centre. There, the data are displayed to the human tele-operator who generates control commands, e.g., desired steering wheel angle or velocity. These are then transmitted back to the vehicle for execution. ToD technology faces a number of challenges that need to be overcome.
Reduced situational awareness poses one of the greatest challenges for ToD, as the tele-operator is not physically located in the vehicle. Additional mental effort is required to compensate for distortions and recreate missing information from the sensor data.
The transmission of signals over mobile networks introduces latency, which can be critical if the vehicle is remotely controlled at stabilization level, i.e., the teleoperator produces direct steering commands. If the latency is too large, different control concepts may be applied, such as an indirect, trajectory-based control scheme. Nevertheless, with 5G technology, nowadays limitations posed by network latency are subject to change.
HD Mapping
One of the cornerstones of autonomous driving is an accurate, actual, and seamless high definition map. The basic functionality is to determine the vehicle’s position — which road and which lane it is in — but also information about traffic rules like speed limitations, or more dynamic conditions like road closures or construction areas. High-definition (HD) map users expect a continuous availability of the map content, even in cross-border scenarios. Autonomous cars, however, require the map to be constantly up-to-date, and thus when reality changes, the map needs to be updated. Regular map updates by the map provider, typically done a few times a year by driving mapping vans along the roads, are not at all sufficient. To ensure a high reliability of autonomous cars, the map needs to be updated constantly, by as many contributing cars as possible. Broadly speaking, the cars collect information about their surroundings using their on-board sensors, and then use their connectivity to send this information to some backend. Here, the received data is compared to the existing map, and if differences are found, the map can be updated. The data might even come from sources other than cars, e.g., road side cameras. The HD map can also be used as the base upon which more dynamic information can be stored, for example, accidents. All these procedures have to work seamlessly across borders. For example, map updates from cars on one side of a border have to be distributed also to cars on the other side, served by a different operator with the backend running on a different Mobile Edge Cloud/Computing (MEC) architecture.
Anticipated Cooperative Collision Avoidance
Towards the realization of autonomous vehicles, car manufacturers are adopting and developing sensors that allow vehicles to sense their environment and control the vehicles. Driving automation systems rely on a variety of sensors like cameras, radar, lidar, etc. Despite the increasing number of in-vehicle sensors, the environmental perception of the vehicle remains limited. In certain situations, typical stand-alone sensing systems will not be able to detect and localize dangerous events on the road with sufficient level of anticipations. In such situations, too late detection of a dangerous event will trigger a hard braking or a dangerous manoeuvre or potentially lead to a collision.
The Anticipated Cooperative Collision Avoidance (ACCA) use case relates to the possibility to anticipate certain potentially critical events in order to reduce the probability of collisions in situations when typical sensors will have no visibility or a short detection range (e.g., a few 100 m). The aim of the ACCA use case is to induce smoother and more homogeneous vehicle reactions by facilitating the anticipated detection and localization of temporarily static events such as traffic jams, high deceleration, emergency braking or unexpected manoeuvres of vehicles in front, etc.